how did our language derive from nothing?

To the extent that there is a mystery, it should be with spoken language, rather than written language. Written language is, after all, just a technology that encodes the spoken language in marks. Spoken language is far older than written language.Humans certainly have a natural ability for language. It is known that a group of deaf children, if not raised with a sign language, will spontaneously invent their own sign language. And if a group of normal children is raised without spoken language -- perhaps they are children of deaf parents who only use sign language -- then these children will spontaneously invent their own spoken language.There is a debate within the field of linguistics, on the nature of language. The majority view is based on that of linguist Noam Chomsky, who believes that language requires a special innate mental organ, which Chomsky refers to as "universal grammar". The classic reference for this is Chomsky's 1972 book "Language and Mind". Steven Pinker is a proponent of the Chomsky view, and believes that language is a result of evolution. Pinker discusses this in his 1994 book "The Language Instinct."The minority view, which I share, is that language is a social adaptation that evolved to support communication between members of a society. While there are certainly brain regions specialized for language, particularly Broca's area, the minority view is that language ability mainly depends on general purpose learning abilities. Philosopher Hilary Putnam is one of the advocates of this position.Note: I don't want to take this thread off-topic into a debate about Chomsky's linguistic theories. I'm just providing a little background.

I'm not sure where I found that. I will retract that assertion for the moment, but I will try to find a source and post that later.For the case of children inventing a sign language, this can be found in the book "Seeing Voices" by Oliver Sacks. See, also Page Not Found: 404 Not Found - Incidently, identical twins will often invent their own language extensions. This one I know from a former colleague whose children were identical twins. She observed this in her own children, and after some inquiring she found that it often happened.

I read it somewhere, but I'm not sure where. Right. Language acquisition has to start relatively early, or there will be problems.

The problems of feral children is generally described as evidence that there is a critical period during which language must be acquired. If language acquisition is delayed beyond that critical period, then it will never work out. What you say about language areas is one possible explanation of that critical period.The Chomskyan view is that language depends on an innate grammar facility (universal grammar), and that this mental organ fits itself to the community language as a maturation process during the child's development. The difficulty for feral children, according to this account, is that by the time they are exposed to language they have passed the maturation phase.My own non-standard view is rather different. I see the grammatical problems as secondary. People can manage to communicate when they flub the grammar. As I see it, the problem is in developing a system of concepts that the child will use to understand its world. For a normal child, many of the important concepts are tied up with social interactions. The feral child has a very different experience, and is not exposed to the same social concepts. When, later, the feral child is introduced to ordinary society, the child's hierarchy of concepts has already been built and it is difficult for that hierarchy to adapt to the conceptual needs of normal society. The child cannot learn to talk about what he cannot properly conceptualize. Yes, that's correct. This seems to be a different phenomenon. There are many natural experiments where people are placed in a culture and language community different from their first language. They generally aquire a pidgin -- a modified form of the second language with somewhat broken grammar. But they don't have the problems that feral children have in adapting to the new culture and language. Derek Bickerton has investigated this, and a web search should turn up a lot of information.

I agree with that. That's how I see it.

Some children do naturally grow up in bilingual homes (the parents speak two languages). It is my understanding that such children pick up both languages quite well. There are probably web pages on it.Hmm, here are some links from searching for bilingual children on google:
http://www.ling.ed.ac.uk/linguist/ask-ling/biling2.html
Multilingual Children's Association
Sorry! That Page Cannot Be Found
Rosenberg-Raising Bilingual Children (TESL/TEFL)

That's what the Chomskyan linguists argue.I don't know your background, so I don't know where you were during the Rubik's cube mania of around 23 years ago. It was clear that children picked it up much more readily than adults. So, by the kind of argument you are using, we should conclude that there is a specific internal brain mechanism for handling a Rubik's cube. I guess we all inherited a Rubik's cube gene. I wonder where that came from.Surely the simpler explanation is that general purpose learning abilities also favor children, and that delayed learning results in poorer learning over many spheres (including Rubik's cube).

So the best age for Rubik's cube is older than 4-5. Nobody is suggesting that there is a universal age at which all learning is better. Sorry, that's wrong. Ask a 4-5 year old about grammar concepts, and he won't understand what you are talking about. He may have good proficiency at speaking grammatically, but he knows little of grammar concepts.In any case, grammar is not the whole issue here. A person who learns a second language as an adult may never get the grammar quite right. Maybe he will talk in some kind of pidgin. But, in spite of the broken grammar, the pidgin speaker manages to communicate quite well. In my opinion, the ability of the pidgin speaker to communicate will with broken grammar is already a problem for Chomskyan linguistics. But that's a bit off-topic, so let's not pursue it.When people talk of a critical period of language acquisition, they are talking about a different kind of problem such as exhibited by feral children. These feral children typically can never communicate well in the language, quite unlike those who use the broken grammar of a pidgin. Again, you are making the mistake of looking to a fixed universal learning age. It's just a fact that a 13 year old can learn algebra far better than can a 23 year old. It is something we mathematicians have to deal with. Somebody who didn't learn algebra in high school will never be fully proficient at it, no matter how many university classes in college algebra he attends.

Yes, sure. And the only possible explanation for biological diversity is that a creator waved his magic wand, and the species all poofed into existence.Shame on you! How can you recognize the argument from ignorance, when the creationists use it, yet then go and use a similar argument from ignorance yourself?Fortran has a grammar. C has a grammar. Pascal has a grammar. Java has a grammar. The evidence that English has a grammar is far weaker.I personally don't believe that English (or other natural language) has a grammar. Sure, we talk about grammar, and the grammaticality of sentences. But it is a grammar that comes from our attempts to systematize language. We impose that grammar on the language, and it does not fit very well.Chomsky's book "Syntactic Structures" was published in 1957. It is almost 40 years later, and Chomskyans still have not given us a definitive grammar for the English language.

I did address the point of your post. Your post was based on argument from ignorance, so is fallacious. I also hinted at an alternative explanation. I'll spell it out.1: Natural language is not a grammatical system.2: Linguists invent a grammar, in their attempt to systematize language.3: Then then impose that grammatical structure on language, even though it doesn't fit very well.4: A child who wants to understand the concepts of this grammar, must first study the systematization. This takes time, and is harder than actually learning the language itself.

Since you are raising that issue, maybe you should tell us how you define grammar. You are presupposing that there is such machinery. What is the basis for that assumption?I'm not denying that humans can handle languages. I am questioning whether there is something that could be called "machinery" involved.Here are some of the reasons I question Chomskyan assumptions:

1. People manage to communicate quite well when using broken grammar. Thus the grammar does not seem to be a necessity.
2. Neuroscientists have not yet discovered this grammar machinery.
3. Chomskyans talk of "generative grammar". AI/Natural Language research, over a period of 50 years, has not been able to design a generative system that resembles natural language.
4. If the brain contains a grammar engine, then this would somehow need to be encoded in the genome. It is doubtful that there is enough DNA in the genome to encode a full specification. But it is worse than that. According to Chomskyans, chimpanzees do not have a grammar engine, but humans do. So the grammar engine would have to be encoded in the parts of the DNA that differ between humans and chimps.
5. There are multiple languages in the world. If there were a grammar engine, you would think that this engine would fix the language, and we would have only one universal natural language.
6. Natural language seems to be mostly a semantic system, and the grammar (if there is one) seems of secondary importance. Yet a Chomskyan style grammatical account of language does not account for semantics.
7. Chomskyans distinguish between performance and capacity. The grammatical theories are said to apply to capacity, not to performance. However, all empirical evidence on how people use language has to do with performance. This distinction seems like a sleight-of-hand move that makes the theory immune to contrary evidence (unfalsifiable).

I use the common terminology, in order to minimize confusion. In my opinion, what we really have is an ad hoc communication protocol, and what is called "grammar" is mostly a side effect of following a protocol. Because a lot of it is not grammar. Here is an example:Kuhl et al., "Linguistic Experience Alters Phonetic Perception in Infants by 6 Months of Age,", Science 255 (1992), pp. 606-608. They learn them at school, in the grammar class Without schooling, I think most kids never learn grammar rules, and never need to learn them. Yet they will still learn to speak in ways that are considered adequately grammatical.

Cultural imperialism

There is a trick to it. Let's consider a hypothetical case, to see how it works.Suppose that a group of professors invent a new language, with what they see as an especially desirable grammar. Then they form a community where they speak only that language. They do this to demonstrate the superiority of the grammar they have designed.Let's call this language NL1 (new language one).As time passes, the people of the community have children. And the children grow up hearing NL1 spoken. So naturally, NL1 is the language that these children try to acquire. And the do acquire it reasonably well, except that they flub the grammar.Let's call the children's language, with flubbed grammar, NL2.As these speakers of NL2 grow up, they eventually have children who grow up hearing some NL1 and a lot of NL2. They do their best, but flub the grammar.Let's call this grandchild language NL3. It is closer to NL2 than NL2 is to NL2.As the NL3 children grow up and have their own children, then in turn acquire a language. And it looks pretty close to NL3.So sure, the language continues to evolve (as all natural languages do). But after a couple of generations, it settles down to a language that children can easily acquire.

I'm not claiming to be able to prove a negative. At present, I don't see that a hypothesized grammar organ explains anything, and I can find no evidence that there is such.